View
218
Download
1
Category
Preview:
Citation preview
Molecular Cell
Article
NEMO Ensures Signaling Specificityof the Pleiotropic IKKb by DirectingIts Kinase Activity toward IkBaBarbel Schrofelbauer,1,2 Smarajit Polley,2 Marcelo Behar,1,2 Gourisankar Ghosh,2 and Alexander Hoffmann1,2,*1Signaling Systems Laboratory2Department of Chemistry and Biochemistry
University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0375, USA*Correspondence: ahoffmann@ucsd.edu
DOI 10.1016/j.molcel.2012.04.020
SUMMARY
Besides activating NFkB by phosphorylating IkBs,IKKa/IKKb kinases are also involved in regulatingmetabolic insulin signaling, the mTOR pathway,Wnt signaling, and autophagy. How IKKb enzymaticactivity is targeted to stimulus-specific substrateshas remained unclear. We show here that NEMO,known to be essential for IKKb activation by inflam-matory stimuli, is also a specificity factor that directsIKKb activity toward IkBa. Physical interaction andfunctional competition studies with mutant NEMOand IkB proteins indicate that NEMO functions asa scaffold to recruit IkBa to IKKb. Interestingly,expression of NEMO mutants that allow for IKKbactivation by the cytokine IL-1, but fail to recruitIkBs, results in hyperphosphorylation of alternativeIKKb substrates. Furthermore IKK’s function in au-tophagy, which is independent of NFkB, is signifi-cantly enhanced without NEMO as IkB scaffold.Our work establishes a role for scaffolds such asNEMO in determining stimulus-specific signal trans-duction via the pleiotropic signaling hub IKK.
INTRODUCTION
Reliable cellular signal transduction depends on the specificity of
kinases. Unlike some metabolic enzymes, in which the catalytic
site may show great specificity for particular small molecule
substrates, protein kinases require specificity domains. When
such specificity domains are encoded by distinct specificity
factors, the kinase activity can in principle be directed to alter-
nate specific substrates, allowing for functional pleiotropy (Bhat-
tacharyya et al., 2006; Ubersax and Ferrell, 2007). One may
distinguish between catalytic (or ‘‘allosteric’’) specificity factors,
that alter the intrinsic catalytic activity of the kinase toward
a specific substrate, and scaffold (or ‘‘tethering’’) specificity
factors, that enhance otherwise weak interactions between
substrate and kinase (Burack and Shaw, 2000). Whereas the
c-Jun N-terminal kinase (JNK)-interacting protein (JIP1), the
JNK/SAPK-activating protein 1 (JSAP1), and the kinase
suppressor of Ras (KSR) are thought to be in the former cate-
gory, the axin scaffolding proteins target the pleiotropic kinases
GSK3 and CK1 to the b-catenin pathway (Burack and Shaw,
2000; Wu and Pan, 2010; Amit et al., 2002), and the Ste5 scaffold
directs the specificity of yeast MAPK signaling (Schwartz and
Madhani, 2004; van Drogen and Peter, 2002). Indeed, shuffling
of binding pockets of Ste5 by site-directed mutagenesis enables
redirection of signaling specificity (Dueber et al., 2003). Thus,
multivalent scaffold proteinsmay not only direct kinase substrate
specificity but also ensure signaling specificity of a pleiotropic
kinase by coordinating upstream and downstream signaling
axes.
The IkB kinase (IKK) complex is ubiquitously expressed and
functionally pleiotropic (Hayden and Ghosh, 2008; Scheidereit,
2006). Its major catalytic component, IKKb (also known as
IKK2), was purified as the IkB kinase that controls nuclear trans-
location of NFkB to initiate proinflammatory gene transcription
by phosphorylating the canonical IkBs, IkBa, IkBb, and IkBε
(Hayden and Ghosh, 2008; Hoffmann and Baltimore, 2006;
Scheidereit, 2006). However, recent literature suggests that
IKKb also plays critical roles in many other biological processes,
including autophagy, insulin signaling, and DNA damage
responses, by targeting diverse alternative cellular substrates
for phosphorylation (Figure 1A) (reviewed in Chariot [2009] and
Scheidereit [2006]).
The large number of substrates, and the diverse biological
functions regulated by their phosphorylation raise the question
of whether IKKb enzymatic activity may be targeted to distinct
substrates. To ensure fidelity as a signal transducer of diverse
signals, one would expect the activation of the kinase by
upstream pathways to be coordinated with downstream
substrate selection.
RESULTS
Constitutively Active IKKb Requires NEMO for NFkBActivationIn response to the inflammatory cytokine IL-1 IKKb rapidly phos-
phorylates S32/36 of IkBa, thereby triggering its ubiquitin-
dependent proteasomal degradation, while phosphorylation of
the NFkB subunit p65 occurs with delayed kinetics (Figures 1B
and S1A). However, using purified, constitutively active IKKb
Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc. 111
Figure 1. Constitutively Active IKKb Is Not Sufficient for NFkB Acti-
vation, but Requires NEMO
(A) IKK is a pleiotroptic transducer capable of phosphorylating many
substrates involved in different biological functions. IKKb phosphorylates the
tumor suppressor p53 at S362 and S366 to regulate its stability (Xia et al.,
2009), the insulin Receptor (IR) substrate 1 (IRS-1) at S307 resulting in the
termination of metabolic insulin signaling (Gao et al., 2002), the tumor
suppressor Tuberous sclerosis 1 (TSC1) at S487 and S511 to activate the
mTOR pathway, enhance angiogenesis and tumor development (Lee et al.,
2007), b-catenin, a key molecule in Wnt signaling, as well as FOXO3a which
acts downstream of growth factor signaling (PI3K/Akt) (Lamberti et al., 2001).
Induction of autophagy depends on IKKb kinase activity; the relevant
substrates remain to be identified (Comb et al., 2011; Criollo et al., 2010).
(B) Phosphorylation of IkBa (S32/36) and p65 (S536) was measured in fibro-
blasts upon stimulation with 1 ng/ml IL-1b.
(C) In vitro activity of constitutively active HA-IKKbEE purified from transfected
293T cells was measured using FL-IkBa or the activation domain of p65
(p65AD) alone (25 min incubation) or mixed together. Time indicates kinase
reaction time.
(D) Steady-state NFkB activity was measured by EMSA in NEMO-deficient
cells supplemented with indicated constructs, and IKK kinase activity was
determined in an in vitro kinase reaction using FL-IkBa as substrate upon
immunoprecipitation of IKKb with an HA antibody.
(E) Western blot analysis of IkBa phosphorylation and steady-state expression
levels in indicated cells with or without MG132 treatment. See also Figure S1.
Molecular Cell
NEMO Directs IKK to NFkB
(HA- IKKbEE) in an in vitro kinase assay, we find high phosphor-
ylation of both IkBa and p65 proteins with no preference for IkBa
(Figure 1C) suggesting that in vivo IKKb’s preference for IkB
112 Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc.
phosphorylation is regulated by additional factors. In vivo, IKKb
activation in response to inflammatory cytokine and pathogen-
associated molecular patterns (PAMPs) stimulation requires
NEMO, which connects the kinase complex to ubiquitin chains
emanating from receptor-associated signaling complexes
(Israel, 2010). To test whether NEMO may also be determining
the target specificity of the catalytic enzyme, we used a constitu-
tively active form of IKKb generated through mutation of the crit-
ical activation loop serines 177 and 181 to phosphomimetic
glutamates (IKKbEE). Retroviral transduction of wild-type fibro-
blasts with IKKbEE resulted in the expected increase in
steady-state NFkB activity, but similar expression of IKKbEE in
NEMO-deficient cells had only a modest effect (Figure 1D).
However, reconstitution with retrovirally expressed NEMO
restored the strong NFkB DNA binding activity. Strikingly,
in vitro kinase assays with immunoprecipitated IKK using either
FL-IkBa (Figure 1D) or GST-IkBa1-54 (Figure S1B) as substrate
showed that NEMOdid not alter the catalytic IKK activity in these
cells, suggesting that it has a specific role in the NFkB activation
pathway at a step downstream of its known function in IKK acti-
vation. Indeed, we found that IKKbEE caused only weak IkBa
phosphorylation and degradation in unstimulated NEMO-defi-
cient cells, unless NEMO was reconstituted (Figure 1E). Taken
together these data indicate that intrinsic substrate specificity
of IKKb is not sufficient for efficient IkB phosphorylation, but
that NEMO plays a role in targeting IKK activity toward the
IkB-NFkB signaling module.
The ZF of NEMO Is Required for EfficientPhosphorylation of IkBaNEMO’s N-terminal region interacts with IKKb, and the central
CoZi-region allows for dimerization and binding to linear and
K63 polyubiquitin chains, connecting IKK to inflammatory
receptor pathways (Figure 2A) (Israel, 2010). A frameshift muta-
tion that completely removes NEMO’s C-terminal Zinc finger (ZF)
causes incontinentia pigmenti; its molecular function however
remains unclear (Makris et al., 2002). While it is critical for
NFkB activation triggered by genotoxic stress (Huang et al.,
2002), its role in inflammation-induced NFkB activation is cell-
and stimulus-specific. Disruption of the ZF was found to prevent
TNF- but not IL-1b-induced NFkB activation in MEFs (Makris
et al., 2002). A complete defect in cytokine-induced NFkB acti-
vation was observed in human T cells expressing human
C417R NEMO (corresponding to C410R in mouse NEMO)
(Yang et al., 2004), while NFkB activation by LPS was normal
in mouse B cells expressing this mutant (Huang et al., 2002).
More recently the ZF has been implicated in enhancing the
affinity of NEMO for binding to K63 linked polyubiquitin chains,
important for IKK activation by specific stimuli (Laplantine
et al., 2009). We examined the possible role of the NEMO ZF
on NFkB activation in the context of constitutively active IKKb.
We generated a truncation (DC25), a debilitating (C389/393S),
and a neutral (M408S) mutant of the ZF domain. Reconstitution
of NEMO-deficient cells with these constructs resulted in
expression levels comparable to those in wild-type MEFs (Fig-
ure S2A). In cells expressing either WT or the M408S NEMO,
which does not affect the overall structure of the ZF, strong
NFkB activity was detected (Figure 2B), while only modest
Figure 2. The Zinc-Finger of NEMO Is Dispensable for IKK Activation by IL-1 but Necessary for NFkB DNA Binding Activity
(A) Domain structure of mouse NEMO. NEMO contains two coiled-coil domains (CC1 and CC2), a leucine zipper (LZ) motif, and a zinc finger (ZF). Mutations
introduced into the NEMO ZF are highlighted in red and green.
(B) NFkB activity (top) was measured in NEMO-deficient cells expressing IKKbEE and indicated NEMO constructs by EMSA. Expression levels of NEMO are
shown below.
(C) NEMO-associated kinase activity was determined in NEMO-deficient cells expressing indicated mutant NEMO in an in vitro IP-kinase assay upon stimulation
with IL-1 for indicated times.
(D) Western blot analysis of phosphorylation status and total levels of IkBa upon stimulation with IL-1.
(E) NFkB DNA binding activity of NEMO-deficient cells reconstituted with the indicated mutant NEMO constructs was measured by EMSA upon stimulation with
IL-1 (10 ng/ml). See also Figure S2.
Molecular Cell
NEMO Directs IKK to NFkB
nuclear NFkB activity was observed in NEMO-deficient cells.
Strikingly, truncating and debilitating the ZF abolished NEMO’s
ability to reduce IkBa levels (aswell as IkBb and IkBε, Figure S2B)
and to enhance NFkB activity in the presence of IKKbEE (Fig-
ure 2B). These data suggest that the ZF of NEMO is critical for
the active IKK complex to efficiently target IkBs for degradation
in vivo, to allow for NFkB activation, even though ZF mutations
had no effect on in vitro IKK kinase activity (Figure S2C).
To examine NEMO’s role in targeting IKK toward IkBs during
inflammatory signaling, we had to identify conditions in which
NEMO mutants, defective in IkB targeting, allowed for efficient
IKK activation. NEMO-deficient cells reconstituted with NEMO
mutants were exposed to TNF, IL-1b, and LPS. In cells express-
ing WT NEMO and M408S mutant NEMO, all stimuli induced
in vitro kinase activity, although kinase activity induced by TNF
was attenuated in M408S NEMO-expressing cells (Figures 2C,
S2D, andS2E). IL-1band LPSalso induced strong, yet somewhat
weaker kinase activity in cells expressing NEMOwith a truncated
or debilitated ZF (Figures 2C and S2D), though both mutants
were defective in supporting TNFa-induced kinase activity (Fig-
ure S2E), consistent with data reported by Makris et al. (2002).
Thus, the ZF of NEMO appears to be largely dispensable for
the activation of the IKK complex by IL-1b and LPS in fibroblasts,
reflecting stimulus-specific requirements for IKK activation.
If the ZFwere necessary for recruiting the IKK complex to IkBs,
a defect in IkBa phosphorylation and degradation and subse-
quent NFkB activation in cells expressing ZF-mutated NEMO
would be anticipated. As expected, in NEMO�/�cells expressingWT NEMO or a neutral M408S mutant, phosphorylation and
degradation of IkBawere induced by IL-1b (Figure 2D). However,
in cells expressingNEMOwith a truncatedor debilitatedZF, IL-1b
stimulation triggered neither IkBa phosphorylation nor degrada-
tion. Furthermore, NFkB DNA binding activity induced by IL-1b,
LPS, andTNFwasdefectivewith ZFmutantswhile it was strongly
activated in WT and M408S-expressing cells (Figures 2E, S2F,
and S2G). The reduction in kinase activity measured in cells ex-
pressing truncated or C389/93S NEMO could not account for
the observed defect in NFkB activation, as similarly low kinase
activities in WT cells (elicited by lower stimulus concentrations)
allowed for detection of strong NFkB DNA binding ability (Fig-
ure S2H). These data are consistent with results obtained with
constitutively active IKKb and further indicate that NEMO does
indeed play a role in NFkB activation that is distinct from its func-
tion in activating the IKK complex, most likely by facilitating the
recruitment of IkBs to the IKK complex.
NEMO Forms a Complex with IKKb and IkBaNEMO’s apparent role as a specificity factor for IKK may be
mediated by allostery altering the catalytic specificity of the
enzyme or by functioning as a scaffold tethering IkB substrates
toward the catalytic site. Given that in vitro kinase assays did
not reveal differences in catalytic activity, we hypothesized that
Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc. 113
Figure 3. NEMO Forms a Complex with IKKb and IkBa
(A) Schematic of a possible complex between NEMO, IKKb, and IkBa. The N-terminal 100 amino acids of NEMO facilitate interaction with IKKb, while the
C-terminal ZF might allow for interaction with IkBa.
(B) NEMO interacts with IkBa via its C terminus. 293T cells were transfected with nondegradable FLAG-IkBa (FLAG-IkBa AA), HA-IKKb, and WT or mutant
Myc-NEMO constructs. Left panel: Complexes were analyzed upon immunoprecipitation of IkBa by western blotting using indicated antibodies. Right panel:
Expression levels of inputs were analyzed in parallel.
(C) The NEMOC terminus is required for signaling to IkBa. 293T cells were transfected with small amounts of IKKbEE together with indicated NEMOmutants, and
IkBa steady-state levels were analyzed by western blotting.
(D) Amino acid sequence of the IkBa signaling responsive domain. Key amino acids discussed in the text are shown in color.
(E) Phosphorylation and expression of WT and DR mutated IkBa stably expressed in IkBa-deficient cells was analyzed upon stimulation with IL-1 by western
blotting. Loading was adjusted to similar IkBa levels.
(F) Phosphorylation ofWT or DRmutant IkBa by IKKbEEwas analyzed in an in vitro kinase assay usingwild-type or DRmutant GST-IkBa1-54 (left panel) or FL-IkBa
(middle panel) or FL- IkBa with S32/36 mutated to alanine (AA) to control for specificity.
(G) Interaction betweenWT andDR IkBawith NEMO and IKKbwas analyzed by immunoprecipitation of IkBawith an a-FLAG antibody from 293T cells followed by
western blot analysis with indicated antibodies.
(H) Gel filtration of recombinant FL-human IkBa, wild-type, or 1–319 truncated human NEMO. Proteins were mixed at a ratio of 1:1 prior to gel filtration. Fractions
13–35 were analyzed by western blotting. See also Figure S3.
Molecular Cell
NEMO Directs IKK to NFkB
NEMO functions as a recruitment scaffold by enhancing the
IKK-IkB interaction through an N-terminal interaction with IKKb
and a C-terminal interaction with IkBa (Figure 3A). To test this
hypothesis, IkBa was immunoprecipitated and its interaction
with IKKb and NEMO was analyzed by immunoblotting (Fig-
ure 3B). A strong interaction between IkB and IKK could be
detected in the presence of WT NEMO, but not in its absence.
Deletion of the ZF abolished binding. The interaction was depen-
dent on IKKb, as the interaction of IkB with WT NEMO was
114 Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc.
reduced in the absence of IKKb or when an N-terminally trun-
cated NEMO, which lacks the IKKb binding site, was used.
Comparable results were obtained when NEMO was immuno-
precipitated (Figure S3). Together, these data indicate that
NEMO interacts with IkBa, that the ZF is required for this interac-
tion, and that the interaction is strengthened in the presence of
IKKb, suggesting the existence of a IkB:IKK:NEMO complex.
In a functional assay, in which steady-state expression of IkBa
was analyzed in the presence of small amounts of IKKbEE,
Figure 4. NEMO Functions as a Scaffold
In Vivo and In Vitro
(A) Schematic and computational model simula-
tion to illustrate combinatorial inhibition and its
consequence on NEMO:IkB:IKK complex forma-
tion and IkBa phosphorylation.
(B) 293T cells were transfected with HA-IKKbEE
and increasing amounts of WT or mutant NEMO
constructs. Levels of phosphorylated and total
IkBa were analyzed by western blotting.
(C) Quantification of IkBa phosphorylation in (B).
p-IkBa was normalized to total IkBa expression.
(D) In vitro kinase assay from 293T cells trans-
fected with HA-IKKbEE and indicated amounts of
WT or mutant NEMO constructs. FL-IkBa was
used as substrate.
(E) In vitro kinase assay using HA-IKKbEE purified
from transfected 293T cells using 50 ng of full-
length wild-type or DR IkBa as substrate in the
presence of 10 mg BSA as nonspecific competitor.
50 or 250 ng recombinant human wild-type of
1–319 NEMO were added to the reaction as indi-
cated. Time indicates kinase reaction time. Inputs
were analyzed by western blotting in parallel. See
also Figure S4.
Molecular Cell
NEMO Directs IKK to NFkB
reduced levels of IkBa could only be detected when WT NEMO
was cotransfected, but not when N- and C-terminally truncated
NEMO was expressed, further suggesting that NEMO indeed
enhances IkBa turnover by means of complex formation with
IkBa and IKKb (Figure 3C).
The N-terminal 54 amino acids of IkBa are known to be suffi-
cient for phosphorylation by IKKb (Figure 3D) (Wu and Ghosh,
2003). As NEMO appears to be important for targeting IKKb to
IkBa, IkBb, and IkBε, we hypothesized that a region conserved
in canonical IkBs would mediate the interaction with NEMO.
Most mutations of conserved residues in IkBa showed no
defects in phosphorylation and degradation upon stimulation
with IL-1b (data not shown). However, mutation of negatively
charged residues D27 and D28 (corresponding to D14 and E15
in IkBb) to positively charged arginines resulted in a strong
reduction of inducible phosphorylation and degradation upon
stimulation with IL-1b (Figure 3E), indicating that these amino
acids are critical for phosphorylation by IKKb in cells. IKKbEE-
induced phosphorylation of both the full-length and 1–54 IkB
D27/28Rmutantswas comparablewithWT IkBa in in vitro kinase
assays (Figure 3F), indicating that the consensus phosphoryla-
tion site for IKKb was not disrupted by the mutation. Instead,
the in vivo defect in phosphorylation may be due to impaired
recruitment by NEMO as the D27/28R IkBa mutation weakened
IkBa’s interaction with NEMO and IKK2 (Figure 3G). Together,
these data indicate that in cells, NEMO forms a complex with
Molecular Cell 47, 111–
IkBa and IKKb and that amino acids D27
and D28 in the N terminus of IkBa are crit-
ical for complex formation.
To test the interaction between NEMO
and IkBa more directly, we produced
highly purified recombinant, bacterially
expressed NEMO and IkBa and per-
formed gel filtration analysis. As shown in Figure 3H, free IkBa
was detected in the low molecular weight fractions 26–35. In
the presence of full-length NEMO IkBa was also present in
NEMO-containing high molecular weight fractions (13–18).
In contrast, no change in the elution profile of IkBawas detected
in the presence of a C-terminal NEMO deletion mutant (lower
panel). These data are indicative of a direct interaction between
NEMO and IkBa that depends on NEMO’s C terminus.
NEMO Functions as a Scaffold for IKKbA hallmark of scaffold proteins is a nonmonotonic dose response
curve: increasing amounts of scaffold allow for increased
complex formation, but when the scaffold concentration
exceeds that of the kinase or substrate whose interaction it coor-
dinates, then the excess will inhibit the formation of the full
complex, by instead favoring the formation of many incomplete
subcomplexes, an effect referred to as ‘‘combinatorial inhibition’’
(Ferrell, 2000; Levchenko et al., 2000). Accordingly, computa-
tional simulations of a mathematical model for the formation of
the IkB-NEMO-IKK complex (Supplemental Experimental
Procedures) showed increased complex formation when
NEMO amounts are increased within a substoichiometric
regime, but reduced functional complexeswithin a higher regime
(Figure 4A). To test this experimentally, we transfected 293T cells
with increasing amounts of NEMO and limiting amounts of
IKKbEE. Expression of small to intermediate amounts of NEMO
121, July 13, 2012 ª2012 Elsevier Inc. 115
Figure 5. Substrate Competition between IkBa, p65, and p105
(A) A model for competitive complex formation between IKK and IkB or alternative substrates. The presence of NEMO favors IKK:IkB complex formation, here
modeled as NEMO-dependent increase of kONIkB.
(B) In silicomodel for IkBphosphorylation. Fraction of IKK bound to IkB and alternative substrate S (right and left, respectively) as a function of the NEMO-induced
stabilization of the IKK-IkB complex (x axis) and the relative concentration of alternative substrate S (y axis). The ‘‘in vitro’’ region is characterized by a very low
ratio of S to IkB in contrast to the ‘‘in vivo’’ situation, in which large amounts of alternative substrates S might be present.
(C) NEMO-deficient cells were stably reconstituted with indicated vectors and phosphorylation of p65 at S536 was analyzed by western blotting. Quantification of
steady-state phosphorylation of IkBa (Figure 1E), p105 (Figure S5A) and p65 (left panel). Phosphorylation signals were normalized to total protein and phos-
phorylation in the absence of NEMO was set to 1.
(D) In vitro kinase assay of HA-IKKbEE using 50 ng of FL-IkBa, FL-p105, or p65AD as substrate in the presence of 10 mg BSA. Where indicated, 250 ng
recombinant WT NEMO were added. Inputs were analyzed by western blotting.
(E) Quantification of phosphorylation of IkBa, p105, and p65 from kinase assays with competitor and limited substrate (D) and without competitor and access
substrate (Figure S5C) after a 15 min kinase reaction. Phosphorylation in the absence of NEMO was set to 1. See also Figure S5.
Molecular Cell
NEMO Directs IKK to NFkB
led to a dose-dependent increase of IkB phosphorylation and
reduced levels of total IkBa expression (Figure 4B and quantifi-
cation Figure 4C) caused by intrinsic IKKbEE activity. Strikingly,
high overexpression resulted in reduced phosphorylation of IkBa
and increased IkBa levels. Expression of NEMO mutants that
only interacted with either IKKb (NEMO DC25) or IkBa (NEMO
DN) had no effect. The kinase activity, assayed in vitro, remained
unchanged throughout all NEMO expression levels (Figures 4D
and S4). Thus, in the context of physical interactions delineated
in Figures 3B–3H, the functional characteristics described in
Figures 4A and 4B support the conclusion that NEMO functions
as a specificity scaffold that recruits classical IkBs to IKK.
Throughout these studies of NEMO’s role as a specificity
factor in vivo, we failed to observe such a function in vitro. We
considered the explanation that the substrate availability/
concentration in cells is lower than in in vitro kinase assay reac-
tion conditions, in which substrate is supplied in excess to
ensure sensitivity and linear dose responses. To test this hypoth-
esis we attempted to mimic in vivo conditions by performing the
kinase reaction in the presence of high concentrations of
116 Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc.
nonspecific competitor protein (BSA) and limiting concentrations
of substrate. Using these conditions, IkBa phosphorylation was
stronger and occurred with faster kinetics in the presence than in
the absence of NEMO (Figure 4E). In contrast, ZF-deleted NEMO
failed to enhance IkBa phosphorylation. Similarly NEMO had
no effect on phosphorylation of IkBa DR, which is phosphoryla-
tion defective in vivo. These data further support the notion
that NEMO acts as a scaffold for IKKb likely by enhancing the
local concentration of IkB to allow for its more efficient
phosphorylation.
NEMO Directs IKKb Activity Away from AlternativeSubstratesIKKb is a pleiotropic kinase that is known to phosphorylate
numerous alternative substrates in vivo, in addition to IkBa. To
analyze the potential effect of the scaffolding function of
NEMO in this in vivo scenario we constructed an in silico model
for IkB phosphorylation, in which IKKb can bind to and phos-
phorylate IkBs or alternative substrates (Figure 5A). Akin to our
experimental results (Figure 4E), model simulations suggested
Molecular Cell
NEMO Directs IKK to NFkB
that in the absence of alternative substrates (i.e., in vitro) the
ZF-dependent NEMO scaffold function would have little effect
on phosphorylation of IkB, while in their presence (i.e., in vivo),
efficient phosphorylation of IkB may only occur in the presence
of WT but not ZF mutant NEMO (Figure 5B, left panel). In
contrast, NEMO’s scaffolding function was predicted to restrict
the phosphorylation of alternate substrates with the ZF mutant
NEMO leading to the hyperphosphorylation of alternative IKKb
substrates (Figure 5B, right panel).
To test this prediction experimentally, we analyzed steady-
state phosphorylation of RelA/p65 Ser536, a well-established
alternative target of IKKb. Only low levels of p65 were phosphor-
ylated in NEMO-deficient cells, but IKKbEE expression greatly
enhanced phosphorylation levels (Figure 5C). Strikingly, expres-
sion of WT NEMO resulted in a strong reduction of phosphory-
lated p65, while ZF-mutated NEMO showed similar p-p65 levels
to the parental NEMO-deficient IKKbEE cells, indicating that in
contrast to IkBa, p65 is a NEMO-independent substrate. Indeed,
in kinase assays performed in the presence of competitor protein
and limiting amounts of substrate to reflect in vivo conditions,
NEMO had no effect on the phosphorylation of p65 or p105,
while IkBa phosphorylation was strongly enhanced by the addi-
tion of NEMO (Figures 5D and 5E). In contrast, without compet-
itor, phosphorylation of IkBa, p65, and p105 was unaffected by
NEMO (Figures 5E and S5C). These data not only demonstrate
that NEMO does not affect IKKb kinase activity per se, but that
it specifically channels IKKb kinase activity to IkBa in conditions
of low substrate abundance.
Next we asked whether we could observe similar substrate
competition in the context of stimulated cells. As expected,
IL-1b stimulation rapidly induced IkBa phosphorylation in
NEMO-deficient cells reconstituted with WT NEMO, (Figure 6A).
No IkBa phosphorylation was detectable in the presence of ZF
mutant NEMO but instead alternative IKKb substrates, p65 and
p105, were strongly phosphorylated. Similar results were ob-
tained upon stimulation with LPS (Figure S6A). Phosphorylation
of p65 and p105 was indeed caused by IKKb, as treatment
with an IKKb-specific inhibitor abolished phosphorylation (Fig-
ure S6B). Despite hyperphosphorylation of p65 and p105 we
did not detect measurable levels of NFkB activity (Figure 2E
and data not shown). These data confirm that NEMO not only
enhances phosphorylation of IkBs, but also restricts the phos-
phorylation of alternative substrates.
We next addressed the implicit hypothesis that IkB and alter-
native substrates are effectively competing in vivo for limited
enzyme activity. Using cells that lack the IkB substrates
(ikba�/�b�/�ε�/�), phosphorylation of both p65 and p105 was
strongly induced as early as 5 min post-IL-1b stimulation, while
it was barely induced inWT cells (Figure 6B). These data suggest
that NEMO’s role as a specificity factor is dependent on the
availability of IkB substrates, further supporting a role of NEMO
in acting as a scaffold/tethering factor to direct IKK’s activity
specifically to IkBs.
Recent studies have shown that IkBa phosphorylation and
degradation are not required for IKKb-induced autophagy, sug-
gesting the involvement of alternative IKKb substrates (Comb
et al., 2011; Criollo et al., 2010). Our model of NEMO as a spec-
ificity factor for IkBs that restricts the phosphorylation of alterna-
tive IKKb substrates would predict that autophagy may be
hyperactivated in cells that express ZF-debilitated NEMO.
Indeed, when ZF-deleted NEMO was expressed along with
IKKbEE, the number of cells that stained positive for the autoph-
agy marker LC3 was significantly enhanced over that of WT
NEMO controls (Figure 6C). Concomitantly, phosphorylation of
p70S6K, a marker for mTOR activity, was strongly reduced,
while the upstream AMPK was hyperphosphorylated in cells ex-
pressing the ZF mutant NEMO (Figure 6D), indicating that the
signaling required for the induction of autophagy is hyperacti-
vated by constitutively active IKKb when NEMOs IkB targeting
function is lost. Similarly, starvation increased the number of
LC3 positive cells harboring a ZF-deleted NEMO (Figure 6E),
despite comparable levels of NEMO-associated IKK kinase
activity (Figure S6C). These data further show that autophagy
is hyperactivated when IKKb loses its ability to phosphorylate
IkBs, but instead hyperphosphorylates alternative substrates.
DISCUSSION
We have presented several lines of evidence that NEMO plays
an essential role in targeting IKK to the IkB-NFkB signaling
module. Previous studies focused on NEMO’s essential role in
the activation of IKK via its ability to bind ubiquitin chains, which
are a hallmark of inflammatory signaling by receptors of the
TNFR and TLR/IL1R superfamilies. Using a constitutively active
IKK variant (Figure 1) and identifying a specific mutation in
NEMO (Figure 2), we were able to distinguish between these
two essential functions in activation and targeting. Physical
interaction studies indicated that IKK-IkB interactions are
enhanced by NEMO, which interacts not only with IKK through
its N terminus (May et al., 2000; Mercurio et al., 1999) but may
also form a stable complex with IkBa via its C terminus (Figure 3).
Using combined in silico kinetic modeling and in vivo experi-
mental analysis we showed that the dose-response characteris-
tics of NEMO are indeed consistent with bivalent interactions
(Figure 4) that support a scaffolding model for NEMO. Consis-
tent with such a model, NEMO enhances IkB phosphorylation
in cells, and in in vitro conditions in which alternative substrates
are abundant (Figure 5). NEMO reduces alternate substrate
phosphorylation (Figure 6) and channels IKK activity specifically
to IkBs; it therefore not only sensitizes the inflammatory pathway
but may also increase its responsiveness to transient and
dynamically regulated IKK signals (Cheong et al., 2006; Werner
et al., 2005).
NFkB activation occurs over a wide dynamic range. In the
cyclin-dependent kinase system, alternative substrates are
thought to function as competitive inhibitors for specific phos-
phorylation (Kim and Ferrell, 2007). The resulting ultrasensitivity
renders the pathway insensitive to weak signals and responsive
only to high levels of stimulus. The role of NEMO may thus be
particularly important at low levels of stimulus, as it counteracts
the ultrasensitivity brought by an abundance of alternative
substrates. However, when IKK is highly activated, NEMO’s
scaffolding function may be less critical. This may in part explain
why Huang et al. (2002) did not observe a strong defect in NFkB
activation in a ZF mutant B cell line in response to high doses
of LPS.
Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc. 117
Figure 6. In the Absence of NEMO Scaffolding, IKKb Hyperphosphorylates Alternative Substrates and Hyperactivates Autophagy
(A) Phosphorylation of p105 (S933), p65 (S536), and IkBa (S32) was measured in NEMO-deficient cells reconstituted with WT or C389/93S mutant NEMO upon
stimulation with IL-1. The quantification of this blot is shown on the right.
(B) WT MEFs and IkBa�/�b�/�ε�/� cells were stimulated with IL-1 and phosphorylation of p105, p65, and IkBa was analyzed by western blotting.
(C) Hyperactivation of NFkB-independent IKKb-induced autophagy. NEMO-deficient cells stably expressing constitutively active IKKb and WT or ZF deleted
NEMO were stained for LC3 and Hoechst and analyzed by immunofluorescence microscopy. Representative images are shown. Quantification of LC3-positive
cells is shown on the right (% cells with LC3 puncta, mean ± SD, n > 150 cells, *p < 0.05).
(D) Western blot analysis of phosphorylated S6K (T389) and AMPKa (T172) in NEMO-deficient cells stably expressing IKKbEE and indicated NEMO construct.
(E) NEMO-deficient cells reconstituted with WT or mutant NEMO were starved with HBSS for 1 hr prior to immunofluorescence staining and analysis. The
quantification of LC3-positive cells is shown on the right (% cells with LC3 puncta, mean ± SD, n > 150 cells, *p < 0.05). See also Figure S6.
Molecular Cell
NEMO Directs IKK to NFkB
We show a direct interaction between recombinant purified
NEMO and IkBa (Figure 3H). Our coimmunoprecipitation exper-
iments suggest that the NEMO-IkBa interaction is strengthened
by IKKb (Figure 3B), suggesting a stable ternary complex in
which the N terminus of NEMO binds to IKK while its C terminus
facilitates interaction with IkBa, and IKKb itself interacts with the
C terminus of IkBa through its ULD-SDD domains (Xu et al.,
2011). Interestingly, this interaction appears to be important for
IKK’s specificity. Our data suggest that the N-terminal aspartic
acids D27/28 in IkBa are also critical for the efficient phosphor-
118 Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc.
ylation of IkBa on S32/36 in vivo, while its ability to be phosphor-
ylated by IKKb remains intact in vitro. Whether NEMO directly
interacts with the IkBa N terminus or whether the N-terminal
aspartic acids are participating in binding to IKK to form the func-
tional ternary complex remains to be determined. However, our
data support the function of NEMO as a scaffold that enhances
the local concentration of IkBa to allow for its specific and rapid
phosphorylation.
Our work not only adds to our understanding of NEMO’s bio-
logical function, it also provides a framework for understanding
Molecular Cell
NEMO Directs IKK to NFkB
IKK as a pleiotropic signal transducer that is involved in
numerous biological processes and is capable of stimulus-
specific phosphorylation of specific substrates. Although IKK’s
name suggests specificity for IkB phosphorylation, it is NEMO
that determines its substrate specificity and function within the
inflammatory pathway. Akin to how the yeast MAPK Ste11p
mediates distinct signaling axes responsive to mating phero-
mone or hyperosmolarity depending on its association with scaf-
fold proteins Ste5p or Pbs2p respectively (Engstrom et al., 2010;
Schwartz and Madhani, 2004; van Drogen and Peter, 2002),
NEMO defines the inflammatory signaling axis mediated by
IKK, owing to its multivalent interactions with ubiquitin chains,
IKK, and IkB. We would expect that other scaffolds might be
identified for IKK’s alternate functions in autophagy or insulin
signaling. Such scaffolds may either be similarly multivalent,
capable of replacing NEMO and thus insulating the pathway
from inflammatory signaling, or they may be bivalent to function
with NEMO to connect IKK to other upstream activation path-
ways or target IKK to other downstream substrates. The latter
scenarios would provide for crosstalk with the inflammatory
signaling axis, which may underlie the pleiotropic functions of
inflammation itself. On a related note, it is not known whether
the IKK-interacting proteins RAP1 or ELKS, which are essential
for phosphorylation of p65 (Teo et al., 2010) or activation by
genotoxic stress (Wu et al., 2010), respectively, function with
NEMO or in its stead. The use of scaffolds to direct IKK’s target
specificity constitutes a modular strategy to facilitate signaling
specificity while allowing for signaling crosstalk within this
regulatory hub.
The DC25 NEMO mutant described here resembles NEMO
expressed in some patients with incontinentia pigmenti (IP) as
a result of a frameshift mutation (Makris et al., 2002). IP is an
X-linked genetic disease thought to be caused for the most
part by inactivation of NFkB signaling and hypersensitivity of
NEMO-deficient cells to TNF-induced apoptosis (Courtois and
Israel, 2011). Interestingly, in IP mouse models an increase of
inflammatory cytokines including TNF and IL-1b could be
observed in the epidermis at the early onset of the disease
(Makris et al., 2000; Nenci et al., 2006; Pasparakis et al., 2002).
Hyperactivation of autophagy reported here might provide an
explanation for this disease phenotype, as autophagy has
recently been implicated to positively contribute to the biogen-
esis and secretion of inflammatory cytokines such as IL-1b
(Dupont et al., 2011).
To harness the potential of the pleiotropic IKK signal trans-
ducer as a drug target in numerous types of inflammatory
diseases, diabetes, and cancers, it is critical to understand
how its activity is directed to different signaling axes and cellular
functions and the degree to which these are insulated from or
connected with each other. The present work identifies NEMO
as a keymolecule in this systems behavior and provides a frame-
work for future investigations.
EXPERIMENTAL PROCEDURES
Cell Culture and Stimulations
Immortalized NEMO-deficient MEFs were generated via the 3T3 protocol from
embryos derived from NEMO�/� mice (Rudolph et al., 2000). IkBa�/�b�/�ε�/�
cells and IkBa�/� cells were described previously (Basak et al., 2007).
3T3 cells were reconstituted by retroviral transduction with WT or
mutant pBABE.mNEMO.puro, pBABE. IkBa.puro constructs, and/or with
pBABE.IKKbEE.hygro or empty vector controls. Cells were selected with puro-
mycin Hydrochloride (Sigma) and hygromycin B (Invitrogen), respectively, and
expression of the stably expressed constructs was verified by western
blotting. Cells were stimulated with recombinant murine TNFa (Roche), LPS
(Sigma), or murine IL-1b (R&D).
Expression and Purification of Recombinant NEMO and IkBa
His-tagged human recombinant full-length or 1–319 truncated human NEMO
was cloned into pET29 expression vector. pET29.hNEMO WT or 1–319 and
pET15.IkBa were transformed into BL21 cells. Expression was induced with
0.25 mM IPTG at room temperature, and upon sonication proteins were puri-
fied in a buffer containing 50 mM Tris (pH 7.5), 300 mM NaCl, 10% Glycerol,
10 mM Imidazole, 25 mM ZnSO4 (for WT NEMO) using Ni-NTA beads. Proteins
were eluted in buffer containing 250 mM Imidazole and dialyzed.
Size Exclusion Chromatography
For size exclusion chromatography individual proteins were mixed with each
other at a ratio of 1:1 and incubated at room temperature for 1hr to form
complexes prior to size exclusion chromatography. 150 ml of each sample
was loaded onto a Superose 6 10/300 GL column (GE Healthcare) pre-equili-
brated with 25 mM Tris (pH 7.5), 150 mM NaCl, 5% Glycerol, 2 mM DTT and
40 mMZnSO4 and resolved at a flow rate of 0.5 ml/min on an AkTA (GE Health-
care) automated liquid chromatography system. Size fractions of 0.5 ml were
collected and an equal volume aliquot of each fraction was analyzed by
western blotting using relevant antibodies indicated in the respective figures.
Gel-Shift Assay and IKK Kinase Assay
Gel-shift assays were performed as previously described (Basak et al., 2007).
Briefly, nuclear extracts were incubated with 32P-labeled 38 bp spanning
double-stranded oligonucleotides containing two consensus kB sites at
room temperature for 20 min prior to complex separation on a nondenaturing
acrylamide gel. Bands were visualized by autoradiography and quantified
using ImageQuant software. In vitro IKK kinase assays were performed as
described previously (Werner et al., 2005). The IKK complex was immunopre-
cipitated from cytoplasmic extracts using a NEMO-specific antibody (BD
or Santa Cruz Biotechnology) and Protein A-sepharose beads. In NEMO-
deficient cells, IKK kinase complexes were precipitated using an IKKa
antibody (Santa Cruz Biotechnology) or HA antibody (Covance) to pull down
HA-tagged IKKb. Beads were subjected to in vitro kinase reaction containing32P-ATP and 1 mg GST-IkBa1-54, GST-IkBa1-54 D27/28R, full-length wild-type
or D27/28R His-IkBa, full-length p105 or GST-p65AD as substrates. For
competitor experiments 10 mg BSA and 50 ng of substrate and indicated
amounts of recombinant wild-type or 1–319 hNEMOwere added to the kinase
reaction. Upon incubation at 30�C for indicated time, reactions were stopped
by boiling samples in SDS sample buffer. Proteins were resolved by SDS-
PAGE and visualized by autoradiography or western blotting.
Coimmunoprecipitation and Western Blotting
For western blot analysis, whole-cell extracts were prepared using RIPA buffer
supplemented with PMSF, DTT, and phosphatase inhibitors. Samples were
normalized for equal amounts of proteins using a Bradford assay (Biorad).
Phospho-specific antibodies to IkBa (S32/36) p65 (S536), p105 (S933),
p-AMPK (T172) and AMPK, p-p70 S6 Kinase (T389), and p70 S6 Kinase
were from Cell Signaling. For coimmunoprecipitation experiments, 293T cells
were transfected with indicated amounts of pcMyc-NEMO, pcFLAG-IkBa, and
pcHA-IKKb or empty vector. Total amounts of DNA used for transfections were
normalized with empty vector. Cells were lysed in buffer containing 10 mM
HEPES-KOH (pH 7.9), 250 mM NaCl, 1 mM EDTA, 0.5% NP-40, 0.2% Tween
20, 1 mM DTT, 1 mM PMSF, and 10 mMMG132. Lysates were precleared with
protein A/G sepharose beads followed by immunoprecipitation with anti-Myc
(Roche), anti-HA (Covance) or anti-FLAG (Sigma) antibody for 2 hr. Complexes
were collected by addition of 10 ml protein A/G Sepharose for 2 hr followed by
extensive washing in lysis buffer at 37�C. Samples were analyzed on immuno-
blots probed with indicated antibodies.
Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc. 119
Molecular Cell
NEMO Directs IKK to NFkB
Microscopy
Cells were grown on poly-L lysine-coated glass coverslips and where indi-
cated treatedwith HBSS for starvation. Cells were fixedwith 4% formaldehyde
for 15 min, incubated with methanol at �20�C for 10 min followed by blocking
with 10% normal goat serum for 1 hr. Anti-LC3A/B antibody (Cell Signaling)
was used for scoring cells undergoing autophagy. Goat-anti-rabbit Alexa Fluor
488 (Invitrogen) was used as secondary antibody. Nuclei were counterstained
with Hoechst. Images were taken on a Zeiss Axio Observer Z1 microscope.
For quantification, at least 50 cells were counted in three independent exper-
iments. Statistical significance was tested using two-sided student’s t test.
SUPPLEMENTAL INFORMATION
Supplemental Information includes six figures and Supplemental Experimental
Procedures and can be found with this article online at doi:10.1016/j.molcel.
2012.04.020.
ACKNOWLEDGMENTS
B.S. is a Leukemia and Lymphoma Society fellow, and M.B. is a Cancer
Research Institute fellow. The work was supported by NCI RO1 grant
CA141722 and NIGMS P50 GM085763.
Received: June 1, 2011
Revised: January 17, 2012
Accepted: April 16, 2012
Published online: May 24, 2012
REFERENCES
Amit, S., Hatzubai, A., Birman, Y., Andersen, J.S., Ben-Shushan, E., Mann, M.,
Ben-Neriah, Y., and Alkalay, I. (2002). Axin-mediated CKI phosphorylation of
beta-catenin at Ser 45: a molecular switch for the Wnt pathway. Genes Dev.
16, 1066–1076.
Basak, S., Kim, H., Kearns, J.D., Tergaonkar, V., O’Dea, E., Werner, S.L.,
Benedict, C.A., Ware, C.F., Ghosh, G., Verma, I.M., and Hoffmann, A.
(2007). A fourth IkappaB protein within the NF-kappaB signaling module.
Cell 128, 369–381.
Bhattacharyya, R.P., Remenyi, A., Yeh, B.J., and Lim, W.A. (2006). Domains,
motifs, and scaffolds: the role of modular interactions in the evolution and
wiring of cell signaling circuits. Annu. Rev. Biochem. 75, 655–680.
Burack, W.R., and Shaw, A.S. (2000). Signal transduction: hanging on a scaf-
fold. Curr. Opin. Cell Biol. 12, 211–216.
Chariot, A. (2009). The NF-kappaB-independent functions of IKK subunits in
immunity and cancer. Trends Cell Biol. 19, 404–413.
Cheong, R., Bergmann, A., Werner, S.L., Regal, J., Hoffmann, A., and
Levchenko, A. (2006). Transient IkappaB kinase activity mediates temporal
NF-kappaB dynamics in response to a wide range of tumor necrosis factor-
alpha doses. J. Biol. Chem. 281, 2945–2950.
Comb, W.C., Cogswell, P., Sitcheran, R., and Baldwin, A.S. (2011). IKK-
dependent, NF-kB-independent control of autophagic gene expression.
Oncogene 30, 1727–1732.
Courtois, G., and Israel, A. (2011). IKK regulation and human genetics. Curr.
Top. Microbiol. Immunol. 349, 73–95.
Criollo, A., Senovilla, L., Authier, H., Maiuri, M.C., Morselli, E., Vitale, I., Kepp,
O., Tasdemir, E., Galluzzi, L., Shen, S., et al. (2010). The IKK complex contrib-
utes to the induction of autophagy. EMBO J. 29, 619–631.
Dueber, J.E., Yeh, B.J., Chak, K., and Lim, W.A. (2003). Reprogramming
control of an allosteric signaling switch through modular recombination.
Science 301, 1904–1908.
Dupont, N., Jiang, S., Pilli, M., Ornatowski, W., Bhattacharya, D., and Deretic,
V. (2011). Autophagy-based unconventional secretory pathway for extracel-
lular delivery of IL-1b. EMBO J. 30, 4701–4711.
120 Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc.
Engstrom, W., Ward, A., and Moorwood, K. (2010). The role of scaffold
proteins in JNK signalling. Cell Prolif. 43, 56–66.
Ferrell, J.E., Jr. (2000). What do scaffold proteins really do? Sci. STKE
2000, pe1.
Gao, Z., Hwang, D., Bataille, F., Lefevre, M., York, D., Quon, M.J., and Ye, J.
(2002). Serine phosphorylation of insulin receptor substrate 1 by inhibitor
kappa B kinase complex. J. Biol. Chem. 277, 48115–48121.
Hayden, M.S., and Ghosh, S. (2008). Shared principles in NF-kappaB
signaling. Cell 132, 344–362.
Hoffmann, A., and Baltimore, D. (2006). Circuitry of nuclear factor kappaB
signaling. Immunol. Rev. 210, 171–186.
Huang, T.T., Feinberg, S.L., Suryanarayanan, S., andMiyamoto, S. (2002). The
zinc finger domain of NEMO is selectively required for NF-kappa B activation
by UV radiation and topoisomerase inhibitors. Mol. Cell. Biol. 22, 5813–5825.
Israel, A. (2010). The IKK complex, a central regulator of NF-kappaB activation.
Cold Spring Harb Perspect Biol 2, a000158.
Kim, S.Y., and Ferrell, J.E., Jr. (2007). Substrate competition as a source of
ultrasensitivity in the inactivation of Wee1. Cell 128, 1133–1145.
Lamberti, C., Lin, K.M., Yamamoto, Y., Verma, U., Verma, I.M., Byers, S., and
Gaynor, R.B. (2001). Regulation of beta-catenin function by the IkappaB
kinases. J. Biol. Chem. 276, 42276–42286.
Laplantine, E., Fontan, E., Chiaravalli, J., Lopez, T., Lakisic, G., Veron, M.,
Agou, F., and Israel, A. (2009). NEMO specifically recognizes K63-linked
poly-ubiquitin chains through a new bipartite ubiquitin-binding domain.
EMBO J. 28, 2885–2895.
Lee, D.F., Kuo, H.P., Chen, C.T., Hsu, J.M., Chou, C.K., Wei, Y., Sun, H.L., Li,
L.Y., Ping, B., Huang, W.C., et al. (2007). IKK beta suppression of TSC1 links
inflammation and tumor angiogenesis via the mTOR pathway. Cell 130,
440–455.
Levchenko, A., Bruck, J., and Sternberg, P.W. (2000). Scaffold proteins may
biphasically affect the levels of mitogen-activated protein kinase signaling
and reduce its threshold properties. Proc. Natl. Acad. Sci. USA 97, 5818–5823.
Makris, C., Godfrey, V.L., Krahn-Senftleben, G., Takahashi, T., Roberts, J.L.,
Schwarz, T., Feng, L., Johnson, R.S., and Karin, M. (2000). Female mice
heterozygous for IKK gamma/NEMO deficiencies develop a dermatopathy
similar to the human X-linked disorder incontinentia pigmenti. Mol. Cell 5,
969–979.
Makris, C., Roberts, J.L., and Karin, M. (2002). The carboxyl-terminal region of
IkappaB kinase gamma (IKKgamma) is required for full IKK activation. Mol.
Cell. Biol. 22, 6573–6581.
May, M.J., D’Acquisto, F., Madge, L.A., Glockner, J., Pober, J.S., and Ghosh,
S. (2000). Selective inhibition of NF-kappaB activation by a peptide that
blocks the interaction of NEMO with the IkappaB kinase complex. Science
289, 1550–1554.
Mercurio, F., Murray, B.W., Shevchenko, A., Bennett, B.L., Young, D.B., Li,
J.W., Pascual, G., Motiwala, A., Zhu, H., Mann, M., and Manning, A.M.
(1999). IkappaB kinase (IKK)-associated protein 1, a common component of
the heterogeneous IKK complex. Mol. Cell. Biol. 19, 1526–1538.
Nenci, A., Huth, M., Funteh, A., Schmidt-Supprian, M., Bloch, W., Metzger, D.,
Chambon, P., Rajewsky, K., Krieg, T., Haase, I., and Pasparakis, M. (2006).
Skin lesion development in a mouse model of incontinentia pigmenti is
triggered by NEMO deficiency in epidermal keratinocytes and requires TNF
signaling. Hum. Mol. Genet. 15, 531–542.
Pasparakis, M., Courtois, G., Hafner, M., Schmidt-Supprian, M., Nenci, A.,
Toksoy, A., Krampert, M., Goebeler, M., Gillitzer, R., Israel, A., et al. (2002).
TNF-mediated inflammatory skin disease in mice with epidermis-specific
deletion of IKK2. Nature 417, 861–866.
Rudolph, D., Yeh,W.C.,Wakeham, A., Rudolph, B., Nallainathan, D., Potter, J.,
Elia, A.J., and Mak, T.W. (2000). Severe liver degeneration and lack of
NF-kappaB activation in NEMO/IKKgamma-deficient mice. Genes Dev. 14,
854–862.
Scheidereit, C. (2006). IkappaB kinase complexes: gateways to NF-kappaB
activation and transcription. Oncogene 25, 6685–6705.
Molecular Cell
NEMO Directs IKK to NFkB
Schwartz, M.A., and Madhani, H.D. (2004). Principles of MAP kinase signaling
specificity in Saccharomyces cerevisiae. Annu. Rev. Genet. 38, 725–748.
Teo, H., Ghosh, S., Luesch, H., Ghosh, A., Wong, E.T., Malik, N., Orth, A., de
Jesus, P., Perry, A.S., Oliver, J.D., et al. (2010). Telomere-independent Rap1 is
an IKK adaptor and regulates NF-kappaB-dependent gene expression. Nat.
Cell Biol. 12, 758–767.
Ubersax, J.A., and Ferrell, J.E., Jr. (2007). Mechanisms of specificity in protein
phosphorylation. Nat. Rev. Mol. Cell Biol. 8, 530–541.
van Drogen, F., and Peter, M. (2002). MAP kinase cascades: scaffolding signal
specificity. Curr. Biol. 12, R53–R55.
Werner, S.L., Barken, D., and Hoffmann, A. (2005). Stimulus specificity of gene
expression programs determined by temporal control of IKK activity. Science
309, 1857–1861.
Wu, C., and Ghosh, S. (2003). Differential phosphorylation of the signal-
responsive domain of I kappa B alpha and I kappa B beta by I kappa B kinases.
J. Biol. Chem. 278, 31980–31987.
Wu, D., and Pan, W. (2010). GSK3: a multifaceted kinase in Wnt signaling.
Trends Biochem. Sci. 35, 161–168.
Wu, Z.H., Wong, E.T., Shi, Y., Niu, J., Chen, Z., Miyamoto, S., and Tergaonkar,
V. (2010). ATM- and NEMO-dependent ELKS ubiquitination coordinates
TAK1-mediated IKK activation in response to genotoxic stress. Mol. Cell 40,
75–86.
Xia, Y., Padre, R.C., De Mendoza, T.H., Bottero, V., Tergaonkar, V.B., and
Verma, I.M. (2009). Phosphorylation of p53 by IkappaB kinase 2 promotes
its degradation by beta-TrCP. Proc. Natl. Acad. Sci. USA 106, 2629–2634.
Xu, G., Lo, Y.C., Li, Q., Napolitano, G., Wu, X., Jiang, X., Dreano, M., Karin, M.,
and Wu, H. (2011). Crystal structure of inhibitor of kB kinase b. Nature 472,
325–330.
Yang, F., Yamashita, J., Tang, E., Wang, H.L., Guan, K., and Wang, C.Y.
(2004). The zinc finger mutation C417R of I-kappa B kinase gamma impairs
lipopolysaccharide- and TNF-mediated NF-kappa B activation through inhib-
iting phosphorylation of the I-kappa B kinase beta activation loop. J. Immunol.
172, 2446–2452.
Molecular Cell 47, 111–121, July 13, 2012 ª2012 Elsevier Inc. 121
Recommended